3 ms·
Actually, the civilized world has version numbers in the source too: use Moose 2.08; That's valid Perl and will blow up if you have Moose 1.02 installed. O
by autarch 13y ago
Actually, the civilized world has version numbers in the source too:
use Moose 2.08;
That's valid Perl and will blow up if you have Moose 1.02 installed. Of course, it'd be even better if you could be more specific than just "2.08 or greater", but it's still useful as is.
- coolsunglasses 13y agoThat still allows your deps to be stateful and out-of-band of the build. I'd prefer the build to be exclusively aware of version numbers in one place and let the code be version agnostic. I'm not a Perl user, I don't consider running sed on my source code to be a "bonus". Still better than Go though.
- dkulchenko 13y ago> I'm not a Perl user, I don't consider running sed on my source code to be a "bonus". That was completely unnecessary.
- coolsunglasses 13y agoHow else do you deal with all the import invocations in the source code files with outdated versions? M4? Make + C preprocessor? Perl generating Perl? I tried to guess the least-absurd option for updating a library. Please inform the rest of us what the standard tooling for handling this in the Perl community is.
- threedaymonk 13y agoOf course. A true Perl user has no need for sed, when there's "perl -pi -e".
- autarch 13y agoThere are a number of tools in Perl that will scan your code base and automatically generate a dependency list based on what it finds. The one I use is called Dist::Zilla (http://dzil.org/ http://dzil.org/), and it automates the entire packaging and library release process. Again, this is how civilized languages work. I'm sure Perl isn't unique here, of course.